Angiotensin converting enzyme (peptidyldipeptide hydrolase, hereinafter referred to as ACE) occupies a central role in the physiology of hypertension. The enzyme is capable of converting the decapeptide angiotensin I, having the sequence EQU AspArgValTyrIleHisProPheHisLeu
to an octapeptide, angiotensin II, by removal of the carboxyterminal HisLeu. The symbols for various chemical entities are explained in the following table:
Ala=L-alanine PA0 Arg=L-arginine PA0 Asp=L-aspartic acid PA0 &lt;Glu=pyro-L-glutamic acid PA0 Gly=glycine PA0 Hip=Hippuric acid (Benzoyl-glycine) PA0 His=L-histidine PA0 Ile=L-isoleucine PA0 Leu=L-leucine PA0 Phe=L-phenylalanine PA0 Pro=L-proline PA0 .DELTA.Pro=L-3,4-dehydroproline PA0 Ser=L-serine PA0 Trp=L-tryptophan PA0 Tyr=L-tyrosine PA0 Val=L-valine PA0 ACE=Angiotensin converting enzyme PA0 Hepes=N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid
Angiotensin I is formed by the action of the enzyme renin, an endopeptidase found in kidney, other tissues and plasma, on a serum .alpha.-2 globulin.
Blood pressure is affected by certain peptides found in the blood. One of these, angiotensin II, is a powerful pressor (blood pressure elevating) agent. Another, bradykinin, a nonapeptide with the sequence ArgProProGlyPheSerProPheArg is a powerful depressor (blood pressure lowering) agent. In addition to a direct pressor effect, angiotensin II stimulates release of aldosterone which tends to elevate blood pressure by causing retention of extracellular salt and fluids. Angiotensin II is found in measurable amount in the blood of normal humans. However, it is found at elevated concentrations in the blood of patients with renal hypertension.
The level of ACE activity is ordinarily in excess, in both normal and hypertensive humans, of the amount needed to maintain observed levels of angiotensin II. However, it has been found that significant blood pressure lowering is achieved in hypertensive patients by treatment with ACE inhibitors. [Gavras, I., et al., New Engl. J. Med. 291, 817 (1974)].
ACE is a peptidyldipeptide hydrolase. It catalyzes the hydrolysis of the penultimate peptide bond at the C-terminal end of a variety of acylated tripeptides and larger polypeptides having an unblocked .alpha.-carboxyl group. The action of ACE results in hydrolytic cleavage of the penultimate peptide bond from the carboxyl-terminal end yielding as reaction products a dipeptide and a remnant.
The reactivity of the enzyme varies markedly depending on the substrate. At least one type of peptide bond, having the nitrogen supplied by proline, is not hydrolyzed at all. The apparent Michaelis constant (Km) varies from substrate to substrate over several orders of magnitude. For general discussion of the kinetic parameters of enzyme catalyzed reactions, see Lehninger, A., Biochemistry, 2nd Ed., Worth Publishers, Inc., New York, 1975, pp. 189-195. Many peptides which are called inhibitors of the enzymatic conversion of angiotensin I to angiotensin II are in fact substrates having a lower Km than angiotensin I. Such peptides are more properly termed competitive substrates. Examples of competitive substrates include bradykinin, and the peptide BPP.sub.5a (also called SQ20475) from snake venom, whose sequence is &lt;GluLysTrpAlaPro.
Numerous synthetic peptide derivatives have been shown to be ACE inhibitors by Ondetti, et al. in U.S. Pat. No. 3,832,337 issued Aug. 27, 1974.
The role of ACE in the pathogenesis of hypertension has prompted a search for inhibitors of the enzyme that could act as antihypertensive drugs. See for example U.S. Pat. Nos. 3,891,616, 3,947,575, 4,052,511 and 4,053,651. A highly effective inhibitor, with high biological activity when orally administered, is D-3-mercapto-2-methylpropanoyl-L-proline, designated SQ14225, disclosed in U.S. Pat. No. 4,046,889 to Ondetti et al., issued Sept. 6, 1977, and in scientific articles by Cushman, D. W. et al., Biochemistry 16, 5484 (1977), and by Ondetti, M. et al., Science 196, 441 (1977). The inhibitor SQ14225 reportedly has an I.sub.50 value of 2.3.times.10.sup.-8 M. The I.sub.50 value reported by Cushman, et al., supra is the concentration of inhibitor required to produce 50% inhibition of the enzyme under a standard assay system containing substrate at a level substantially above K.sub.m. It will be understood that I.sub.50 values are directly comparable when all potential factors affecting the reaction are kept constant. These factors include the source of enzyme, its purity, the substrate used and its concentration, and the composition of the assay buffer. All I.sub.50 data reported herein have been performed with the same assay system and same enzyme (human urinary ACE) and with an approximately 1/2 K.sub.m level of substrate and are therefore internally consistent. Discrepancies with data obtained by other workers may be observed. Indeed such discrepancies do exist in the literature, for unknown reasons. See, for example, the I.sub.50 values for BPP.sub.9a reported by Cushman, D. W., et al., Experientia 29, 1032 (1973) and by Dorer, F. E., et al., Biochim. Biophys. Acta 429, 220 (1976).
The mode of action of SQ14225 has been based upon a model of the active site of ACE developed by analogy with the better known related enzyme, carboxypeptidase A. The active site was postulated to have a cationic site for binding the carboxyl end group of the substrate and a pocket or cleft capable of binding the side chain of the C-terminal amino acid and providing especially tight binding for the heterocyclic ring of a terminal proline residue. A similar pocket for the penultimate amino acid residue was postulated, and the published data suggested a rather stringent steric requirement, since the D-form of the inhibitor was substantially more potent than its stereoisomer or the 3-methyl and unsubstituted analogs. The sulfhydryl group on the inhibitor, postulated to be bound at the active site near the catalytic center, was believed to play a central role in inactivation of the enzyme by combining with the zinc moiety known to be essential for catalytic activity. Substituents on the sulfhydryl, such as a methyl group, and an S-acetyl derivative, substantially reduced potency of the inhibitor. See Cushman, D. W., et al., Biochemistry, supra.
In vitro study of the mechanism by which SQ14225 and its analogs act to inhibit ACE has been somewhat hampered by the instability of these molecules under ambient conditions. For example, it has been observed that a fresh aqueous solution of concentration, e.g., 1 mg per ml of SQ14225 at a pH of about 8 becomes substantially less active upon standing for as little as 30 minutes, and that activity continues to decrease as the solution stands for longer periods. It is believed that this loss in activity is mainly the result of dimerization of SQ14225 occurring at the sulfhydryl end groups, whereby a disulfide is formed which is largely inactive as an inhibitor. Since the free sulfhydryl group is highly reactive and may be readily oxidized to polar acidic moieties such as sulfone and sulfoxide groups, it may also be that the observed in vitro loss of activity of aqueous solutions of SQ14225 on standing is in some part a consequence of one or more such oxidation reactions, with formation of a sulfone or sulfoxide which does not function effectively as an inhibitor for ACE.
Such reports of SQ14225 clinical testing as are currently available, some of which refer to the compound under the name "Captopril", suggest that the product is sufficiently stable in the normal gastric and intestinal environments of most patients to be an effective inhibitor for ACE when administered orally. It is not yet clear, however, whether there may be a group of patients for which SQ14225 is substantially ineffective. Because of the high reactivity of the free sulfhydryl group, SQ14225 could readily form mixed disulfides with serum, cellular proteins, peptides or other free sulfhydryl group-containing substances in the gastric or intestinal environments, in addition to the possibility for dimer formation or oxidative degradation reactions. A mixed disulfide with protein may be antigenic and, indeed, occasional allergic reactions have been clinically observed. See Gavras, et al., New England J. Med. 298, 991 (1978). Disulfides and oxidative degradation products of SQ14225, if formed, may at best be expected to be largely ineffective as inhibitors. It may be postulated accordingly that dose response to SQ14225 may vary with conditions of administration and among individual patients. Moreover, in at least some patients, unwanted side effects may occur and maintenance of an effective concentration of the inhibitor in the body may be difficult to control.
Thioester compounds generally are thought to be highly reactive in that the thioester linkage is readily hydrolyzable to a sulfhydryl moiety and a carboxylic moiety. Thioesters are accordingly often used as active ester intermediates for acylation under mild conditions. Such groups as, e.g., acetylthio have been used as blocking groups in the above cited Ondetti, et al. patents. Thioester intermediates are also postulated to occur in the biosynthesis of cyclic peptides such as tyrocidin or gramicidin S. See Lipmann, F. in Accounts Chem. Res. 6, 361 (1973).
Thioester compounds having potent ACE inhibitory activity and oral effectiveness as anti-hypertensive agents have been disclosed in copending applications Ser. Nos. 064,897 through 064,903, all filed on Aug. 14, 1979, Ser. Nos. 161,150 and 161,151, both filed on Jan. 30, 1980 and Ser. No. 121,188, filed on Mar. 3, 1980. All copending applications are incorporated herein by reference.
Compounds related to SQ14225 have been disclosed by Ondetti, et al., U.S. Pat. Nos. 4,046,889, 4,052,511, 4,053,651, 4,113,715 and 4,154,840. Of interest are disclosed analogs of SQ14225 having the five-membered heterocyclic ring of proline replaced by a four- or six-membered ring. The inhibitory potencies of such analogs relative to SQ14225 are not disclosed. Substitution of D-proline for L-proline is reported to drastically reduce inhibitory potency of 3-mercaptopropanoyl amino acids (Cushman, D. W., et al., supra).
The substitution of L-3,4-dehydroproline for proline has been studied in several systems. Substitution of L-3,4-.DELTA.Pro in the 7 position of bradykinin yields a bradykinin derivative which has significantly reduced physiological activity. See Fisher, G. H. et al., Arch. Biochem. Biophys. 189, 81 (1978). On the other hand, substitution of L-3,4-.DELTA.Pro at the 3, 5, 8 or 9 position in ACE inhibitor BPP.sub.9a enhances its inhibitory activity. See Fisher, G. H. et al., FEBS Letters 107, 273 (1979). In copending application Ser. No. 161,151, applicants found that the compounds having .DELTA.Pro, which are disclosed in said application, have high inhibitory potency and antihypertensive effectiveness. However, at present, no rationale can be advanced to explain the diversity of observed results following substitution of .DELTA.Pro for proline. Similarly, no clear picture has emerged of the effects of other proline derivatives or analogs substituted at various loci on ACE inhibitors.
To date, the effect of the amino acid to the left of the sulfur in the thioester compounds disclosed in our copending applications, has not been determined. It is thought that this amino acid functions as an additional recognition site for the enzyme. If this is true, it would be expected that a compound with an amino acid here would be a better inhibitor. Applicants have found that various amino acids are effective and that the hydroxyprolines, proline, L-, and D,L-3,4-dehydroproline, thiazolidine-4-carboxylic acid and L-5-oxo-proline derivatives are all effective anti-hypertensive agents and have high inhibitory potency for ACE.